This paper extends the traditional Difference-in-Differences (DID) design to handle multiple time periods and variation in treatment timing. The authors propose a two-step estimation strategy to identify and estimate treatment effect parameters, establish the asymptotic properties of the estimators, and develop a computationally convenient bootstrap procedure for inference. They also introduce a semiparametric data-driven testing procedure to assess the credibility of the DID design. The paper applies these methods to analyze the effect of minimum wage on teen employment from 2001 to 2007, addressing challenges related to variation in state-level minimum wage policy changes. The results suggest that increasing the minimum wage tends to decrease teen employment, with effects ranging from 2.3% to 13.6% lower employment across groups and time. The authors also find evidence against the parallel trends assumption, highlighting the importance of conditioning on observed covariates for valid inference.This paper extends the traditional Difference-in-Differences (DID) design to handle multiple time periods and variation in treatment timing. The authors propose a two-step estimation strategy to identify and estimate treatment effect parameters, establish the asymptotic properties of the estimators, and develop a computationally convenient bootstrap procedure for inference. They also introduce a semiparametric data-driven testing procedure to assess the credibility of the DID design. The paper applies these methods to analyze the effect of minimum wage on teen employment from 2001 to 2007, addressing challenges related to variation in state-level minimum wage policy changes. The results suggest that increasing the minimum wage tends to decrease teen employment, with effects ranging from 2.3% to 13.6% lower employment across groups and time. The authors also find evidence against the parallel trends assumption, highlighting the importance of conditioning on observed covariates for valid inference.